Rice Leaves Disease Detection Mechanism Using VGG16 Deep Learning Architecture

In several nations, agriculture is the main economy as well as source of survival. There are various food crops, in which rice is one of the most favorite grains among people. But rice crop is affected by various diseases at different phases. A variety of diseases cause problems with the quality of the crop and its growth. Because certain diseases show the same symptoms, it can be difficult to diagnose the condition using standard methods or with the naked eye at an early stage. Automation, on the other hand, is extremely useful in detecting diseases at the correct moment in crop production to protect their crops from loss in the initial phases. To develop such automation processes, researchers use various approaches such as image recognition, deep learning, and machine learning. In present work, we applied VGG16 model to suggest a disease classifier to recognize and classify five kinds of rice diseases. Training and validation of proposed model are done on a dataset having five kinds of well-known rice diseases (Brown spot, Blight, Rice Blast, Sheath Blast, and Tungro). 100% validation accuracy is achieved by the suggested model.